Vitamin K antagonists vs. direct oral anticoagulants after transcatheter aortic valve implantation in atrial fibrillation

Aims To examine the risk of arterial thromboembolism, bleeding, and all-cause mortality in atrial fibrillation (AF) patients treated with direct oral anticoagulants (DOACs) vs. vitamin K antagonists (VKAs) undergoing transcatheter aortic valve implantation (TAVI). Methods and results In this nationwide observational cohort study, 735 patients undergoing TAVI from 1 January 2012 to 30 June 2017 with a history of AF and who were treated with oral anticoagulants were identified using data from Danish nationwide registries. Of these, 219 (29.8%) and 516 (70.2%) patients were treated with DOACs and VKAs, respectively. The DOAC group was characterized by a higher prevalence of previous arterial thromboembolism and a lower prevalence of chronic kidney disease compared with the VKA group. The distribution of age, sex, CHA(2)DS(2) -VASc and HAS-BLED scores, and concomitant antiplatelet therapy was similar between groups. Compared with VKA, treatment with DOACs was not associated with a significantly different 3-year absolute risk of arterial thromboembolism [9.6% (95% confidence interval, CI 4.7-16.5%) vs. 7.4% (95% CI 4.9-10.5%) in the DOAC and VKA group, respectively], bleeding [14.3% (95% CI 7.6-22.9%) vs. 13.3% (95% CI 9.9-17.1%)], or all-cause mortality [32.7% (95% CI 21.8-44.0%) vs. 32.0% (95% CI 26.8-37.3%)]. In adjusted analyses, treatment with DOACs, when compared with VKAs, was not associated with a significantly different rate of arterial thromboembolism [hazard ratio (HR) 1.23 (95% CI 0.58-2.59)], bleeding [HR 1.14 (95% CI 0.63-2.06)], or all-cause mortality [HR 0.93 (95% CI 0.61-1.40)]. Conclusion In patients with AF undergoing TAVI, treatment with DOACs was not associated with a significantly different risk of arterial thromboembolism, bleeding, or all-cause mortality compared with VKA.

Automated summary

This study found that in AF patients undergoing TAVI, treatment with DOACs compared to VKAs did not significantly increase the risk of arterial thromboembolism, bleeding, or all-cause mortality.